Androgen ablation remains the standard therapy with advanced prostate cancer (CaP) and causes disease remission in most men. However, CaP eventually recurs and thereafter the median survival of patients is less than one year. Thus the transition from androgen-dependent to androgen- independent growth represents a critical juncture in the progression of prostate cancer. Development and growth of the prostate gland require androgen, a functioning androgen receptor (AR) and expression of growth promoting AR target genes. Expression of AR in recurrent Cap is similar to that in androgen-dependent CaP and our results from studies on the androgen-dependent CWR22 human CaP xenograft indicate that expression of AR-regulated genes is increased in recurrent CWR-22 despite the absence of testicular androgen. The Program Project's unifying hypothesis is that AR and AR-regulated genes have a role in the recurrent growth of CaP during androgen deprivation therapy. The Program consists of three closely related projects. Project 1 will test the hypothesis that changes in the expression and function of AR effect the progression of CaP and that these changes differ between African Americans and Caucasian Americans. AR and tissue levels of androgens will be assayed. Mutations in AR will be identified and the function of mutant Ars characterized. Serial biopsies of CaP spanning the period before and after tumor recurrence following androgen deprivation will be analyzed for AR mutations, AR gene amplification and the expression of AR and AR-induced growth promoting genes Project 2 will test the hypothesis that AR is reactivated in the recurrent CWR-22 by an androgen-independent mechanism and its activation in the absence of androgen increases the expression of AR-regulated growth promoting genes. AR in recurrent CWR-22 will be analyzed biochemically for properties associated with activation. Androgen-regulated growth promoting genes will be identified and their expression in recurrent CWR22 determined. AR will be knocked out in recurrent CWR22 by expression of a dominant negative mutant AR or an anti-AR ribozyme to determine the role of AR in up-regulation of the growth promoting genes. Depending on these results the mechanism of androgen-independent AR activation or AR-independent expression of growth promoting AR target genes will be identified. Project 3 will test the hypothesis that recurrent CaP is the outgrowth of a transformed prostate stem cell that contains AR, can be stimulated to proliferate by androgen, but is insensitive to apoptosis induced androgen deprivation. Stem cell lines will be transformed and their capacity to develop androgen-dependent and androgen-independent CaP determined. Additionally, the causal role of AR in maintenance of tumorigenic potential of androgen-independent CaP will be determined. Tumor stem cells will be isolated from the CWR22 xenograft and the role of AR evaluate in the production of tumors when transplanted into castrated mice. Comparable studies will be undertaken to characterize the phenotype of the stem cell components in transient xenografts established from human CaP and determine their capacity to form androgen-independent tumors. Characterization of tumor stem cell populations will allow analysis of archived specimens for differences in frequencies of stem cells between androgen-dependent and androgen-independent CaP.